Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 51 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.1090. Min Act: -0.0002

Data Index: 279529 (C4 (Web Text))

Max Activating Token Index: 370

Text #1

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0962. Min Act: -0.0002

Data Index: 1648605 (Python Code)

Max Activating Token Index: 438

Text #2

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0835. Min Act: -0.0002

Data Index: 1470001 (Python Code)

Max Activating Token Index: 779

Text #3

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0854. Min Act: -0.0002

Data Index: 1098958 (C4 (Web Text))

Max Activating Token Index: 867

Text #4

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0784. Min Act: -0.0002

Data Index: 1183533 (C4 (Web Text))

Max Activating Token Index: 156

Text #5

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0838. Min Act: -0.0002

Data Index: 670664 (C4 (Web Text))

Max Activating Token Index: 583

Text #6

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0804. Min Act: -0.0002

Data Index: 1237833 (C4 (Web Text))

Max Activating Token Index: 795

Text #7

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0822. Min Act: -0.0002

Data Index: 1529937 (Python Code)

Max Activating Token Index: 915

Text #8

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0701. Min Act: -0.0002

Data Index: 1267126 (C4 (Web Text))

Max Activating Token Index: 205

Text #9

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0777. Min Act: -0.0002

Data Index: 786879 (C4 (Web Text))

Max Activating Token Index: 147

Text #10

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0801. Min Act: -0.0002

Data Index: 782326 (C4 (Web Text))

Max Activating Token Index: 937

Text #11

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0784. Min Act: -0.0002

Data Index: 1360152 (Python Code)

Max Activating Token Index: 766

Text #12

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0829. Min Act: -0.0002

Data Index: 712434 (C4 (Web Text))

Max Activating Token Index: 373

Text #13

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0749. Min Act: -0.0002

Data Index: 622393 (C4 (Web Text))

Max Activating Token Index: 621

Text #14

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0699. Min Act: -0.0002

Data Index: 1383398 (Python Code)

Max Activating Token Index: 923

Text #15

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0702. Min Act: -0.0002

Data Index: 1600011 (Python Code)

Max Activating Token Index: 553

Text #16

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0724. Min Act: -0.0002

Data Index: 1408621 (Python Code)

Max Activating Token Index: 240

Text #17

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0721. Min Act: -0.0002

Data Index: 996911 (C4 (Web Text))

Max Activating Token Index: 895

Text #18

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0818. Min Act: -0.0002

Data Index: 1299131 (C4 (Web Text))

Max Activating Token Index: 950

Text #19

Max Range: 0.1090. Min Range: -0.1090

Max Act: 0.0749. Min Act: -0.0002

Data Index: 394804 (C4 (Web Text))

Max Activating Token Index: 935